Hi everyone,
I am currently working on an op amp buffer (based on the MCP6292 op amp), which must be powered from the Raspberry pi 5V GPIO.
The op amp works as expected, but I am facing some issues :
1 - When I plug everything (Audio IN from Guitar and Audio OUT to amp), no noise can be heard.
2 - When I plug everything and I play a note on my guitar, the sound out to the amp with "switching-noise".
Actually, when I play a note on my guitar, I can hear the Green activity LED blinking on the raspberry pi.
For my point of view, this is definitely a coupling issue, but I can't figure out where this happens.
I attach the schematic I use.
Could someone help me to find issue on my schematic?
I really appreciate.
Thank you.
Marius
I am currently working on an op amp buffer (based on the MCP6292 op amp), which must be powered from the Raspberry pi 5V GPIO.
The op amp works as expected, but I am facing some issues :
1 - When I plug everything (Audio IN from Guitar and Audio OUT to amp), no noise can be heard.
2 - When I plug everything and I play a note on my guitar, the sound out to the amp with "switching-noise".
Actually, when I play a note on my guitar, I can hear the Green activity LED blinking on the raspberry pi.
For my point of view, this is definitely a coupling issue, but I can't figure out where this happens.
I attach the schematic I use.
Could someone help me to find issue on my schematic?
I really appreciate.
Thank you.
Marius
Attachments
You should use some ferrite bead/s and proper decoupling (10uF near the bead/s, and 100nF+1nF right next to the opamp) . The 3.3V generator on the Pi is very, very noisy.
Looks fine, I would add the 10u after the bead (or in addition to before for better rejection to source).
Also, this is a very high input impedance circuit running at a (relatively speaking) higher supply current, so layout is critical. Your input circuit should ideally be right at the opamp pin, and fully separated from the power routing.
It might help to also have some HF filtering with a RC filter at input, like a 1k + 470pF combination. This should lower the circuit bandwidth and make it a little better at noise rejection from cables. Some output termination after the capacitor might also be nice.
Your output is not connected in any way to the Pi now, is it?
Also, this is a very high input impedance circuit running at a (relatively speaking) higher supply current, so layout is critical. Your input circuit should ideally be right at the opamp pin, and fully separated from the power routing.
It might help to also have some HF filtering with a RC filter at input, like a 1k + 470pF combination. This should lower the circuit bandwidth and make it a little better at noise rejection from cables. Some output termination after the capacitor might also be nice.
Your output is not connected in any way to the Pi now, is it?
Thanks for your advice. I will try this.
Your output is not connected in any way to the Pi now, is it?
Actually, this op amp circuit is connected to an IQAudio Codec sound card (https://www.kubii.fr/hat-dac-modules/3183-codec-zero-iqaudio-0617588405556.html). The OUTPUT on my schematic is connected to AUX IN left, and GROUND is connected to AUX IN GROUND.
Another point :
Even if I used a second power supply for the op amp circuit, and if I connect OUTPUT to AUX IN, I still hear the noise (Blinking GREEN activity LED on Rpi).
I don't really understand this point...
Your output is not connected in any way to the Pi now, is it?
Actually, this op amp circuit is connected to an IQAudio Codec sound card (https://www.kubii.fr/hat-dac-modules/3183-codec-zero-iqaudio-0617588405556.html). The OUTPUT on my schematic is connected to AUX IN left, and GROUND is connected to AUX IN GROUND.
Another point :
Even if I used a second power supply for the op amp circuit, and if I connect OUTPUT to AUX IN, I still hear the noise (Blinking GREEN activity LED on Rpi).
I don't really understand this point...
Since you already figured out the ground loop issue, here's a possible solutioon:
https://www.mouser.in/ProductDetail...GAEpiMZZMsc0tfZmXiUnS/tm3ue3cnPKo7QFQut%2Ba8=
In any case, the activity light may be connected to the IQAudio's I2C bus, and may indicate when the card is receiving signal. If you have audible spuriae, it may be basically anything from the very poor power solution on the Pi, to some issues with the IQAudio card. You could try the opamp circuit with a regular guitar amp and see if that is the culprit, else the issues may rest within the Pi + IQAudio system. I use some of the IQAudio DAC HATs' and they are extremely sensitive to ground loops, even having no input circuitry. I had a world of trouble powering the boards properly, and eventually gave up trying.
https://www.mouser.in/ProductDetail...GAEpiMZZMsc0tfZmXiUnS/tm3ue3cnPKo7QFQut%2Ba8=
In any case, the activity light may be connected to the IQAudio's I2C bus, and may indicate when the card is receiving signal. If you have audible spuriae, it may be basically anything from the very poor power solution on the Pi, to some issues with the IQAudio card. You could try the opamp circuit with a regular guitar amp and see if that is the culprit, else the issues may rest within the Pi + IQAudio system. I use some of the IQAudio DAC HATs' and they are extremely sensitive to ground loops, even having no input circuitry. I had a world of trouble powering the boards properly, and eventually gave up trying.
Thanks for the proposed component. I will try it asap.
Regarding the test-bed you proposed, I actually already did that. I have connected the output of the op amp circuit to my guitar amp and there is no noise (at least no 'bad' noise). This was the first test I made to identify the root cause of my problem. As you stated (and what I supposed), the issue is within the Pi + IQaudio cart.
I will try to use an oscilloscope to find out where the ground loop appear on my schematic. I will be back for update asap.
You said 'I had a world of trouble powering the boards properly, and eventually gave up trying'. Does it mean you tried using other RPi audio board? If so, could you tell me what kind of hat?
Thank you again for your help.
Marius
Regarding the test-bed you proposed, I actually already did that. I have connected the output of the op amp circuit to my guitar amp and there is no noise (at least no 'bad' noise). This was the first test I made to identify the root cause of my problem. As you stated (and what I supposed), the issue is within the Pi + IQaudio cart.
I will try to use an oscilloscope to find out where the ground loop appear on my schematic. I will be back for update asap.
You said 'I had a world of trouble powering the boards properly, and eventually gave up trying'. Does it mean you tried using other RPi audio board? If so, could you tell me what kind of hat?
Thank you again for your help.
Marius
I do not use any input boards, but I have a few DAC HATs. I have both the IQAudio units, one is the DAC+ and the other is the DAC Pro ( I think). When running the whole thing from a buck converter connected to the power supply for the amplifier, the buzz was terrible, worse than you might expect from a single looped ground. I also have used the Allo Zero (which is dead quiet), a Chinese Digi clone, and a few ES9018K2M I2S input boards (not exactly a HAT). All of them worked sort of OK, even with a buck converter. They had a bit of noise, but not terrible.
Eventually I moved to USB DACs and the IQAudio HATs are sitting around unemployed. I use it maybe once a year, just to see if it still works.
Eventually I moved to USB DACs and the IQAudio HATs are sitting around unemployed. I use it maybe once a year, just to see if it still works.